Title | ||
---|---|---|
An improved classification of hyperspectral imaging based on spectral signature and gray level co-occurrence matrix. |
Abstract | ||
---|---|---|
Hyperspectral imaging (HSI) has been used to perf orm objects identification and change detection in natural environment. Indeed, H SI provide more detailed information due to the high spectral, spatial and temporal resoluti on. However, the high spatial and spectral resolutions of HSI enable to precisely characterize the information pixel content. In this work, we are interested to improve the classification of H SI. The proposed approach consists essentially of two steps: features extraction and cl assification of this data. Most conventional approaches treat the spatial information without con sidering the spectral information contained in each pixel, for that, we propose a new a pproach for features extraction based on spatial and spectral tri-occurrence matrix defined on cubic neighborhoods. This method enables the integration of the spectral signature i n the classical model for calculating the co- occurrence matrix to result the 3D-Gray Level Co-occ urrence Matrix (GLCM). Concerning the classification step, we are mainly interested in the supervised classification approach. We used the Support Vector Machine (SVM) allowing class ification without using a dimensionality reduction. We will consequently test the proposed approach on a n IHS that was recorded by an AVIRIS sensor. Itu0027s an Indiana Pines scene which is a veget ation zone captures in north-western Indiana. Itu0027s composed of two spatial dimensions of size 145X145 pixels and with spatial resolution of 20m per pixel, and a spectral dimensi on with 220 bands. The choice of this image is melted by the existence of a ground truth and its permanent use in all IHS analysis problems. The experimental results indicate a mean accuracy values of 70.73% for VGLCM. It shown the robustness of our perspective approach better classification rate and high accuracy. |
Year | Venue | Field |
---|---|---|
2015 | SAGEO | Spatial analysis,Computer vision,Dimensionality reduction,Pattern recognition,Co-occurrence matrix,Hyperspectral imaging,Feature extraction,Ground truth,Artificial intelligence,Pixel,Spectral signature,Geography |
DocType | Citations | PageRank |
Conference | 1 | 0.35 |
References | Authors | |
5 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Fedia Ghedass | 1 | 1 | 0.35 |
Imed Riadh Farah | 2 | 86 | 26.16 |